II.: of the conception and the necessity of a metaphysic of ethics.

It has been shown elsewhere that for physical science which has to do with the objects of the external senses we must have à priori principles; and that it is possible—nay, even necessary—to prefix a system of these principles under the name of metaphysical principles of natural philosophy to physics, which is natural philosophy applied to special phenomena of experience. The latter, however (at least when the question is to guard its propositions from error), may assume many principles as universal on the testimony of experience, although the former, if it is to be in the strict sense universal, must be deduced from à priori grounds; just as Newton adopted the principle of the equality of action and reaction as based on experience, and yet extended it to all material nature. The chemists go still further, and base their most universal laws of combination and dissociation of substances by their own forces entirely on experience, and yet they have such confidence in their universality and necessity that, in the experiments they make with them, they have no apprehension of error.

It is otherwise with the moral laws. These are valid as laws only so far as they have an à priori basis and can be seen to be necessary; nay, the concepts and judgments about ourselves and our actions and omissions have no moral significance at all, if they contain only what can be learned from experience; and should one be so misled as to make into a moral principle anything derived from this source, he would be in danger of the grossest and most pernicious errors.

If the science of morals were nothing but the science of happiness, it would be unsuitable to look out for à priori principles on which to rest it. For however plausible it may sound to say that reason could discern, even before experience, by what means one might attain a lasting enjoyment of the true pleasures of life, yet everything which is taught on this subject à priori is either tautological or assumed without any foundation. It is experience alone that can teach us what gives us pleasure . The natural impulses to nutrition, to the propagation of the species, the desire of rest, of motion, and (in the development of our natural capacities) the desire of honour, of knowledge, &c., can alone teach, and moreover teach each individual in his own special way, in what to place those pleasures; and it is these also that can teach him the means by which he must seek them. All plausible à priori reasoning is here at bottom nothing but experience raised to generality by induction: a generality, too, so meagre that everyone must be allowed many exceptions, in order to make the choice of his mode of life suitable to his special inclination and his susceptibility for pleasure; so that after all he must become wise only by his own or others’ loss. It is not so with the doctrines of morality. They are imperative for everyone without regard to his inclinations, solely because and so far as he is free, and has practical reason. Instruction in its laws is not drawn from observation of himself and his animal part; not from perception of the course of the world, from that which happens and from the way in which men act (although the German word “sitten,” like the Latin mores, signifies only manners and mode of life); but reason commands how men should act, even although no instance of such action could be found; moreover, it pays no regard to the advantage which we may hereby attain, which certainly can only be learned by experience. For although it allows us to seek our advantage in every way that we can; and in addition, pointing to the testimony of experience, can promise us, probably and on the whole, greater advantages from following its commands than from transgression of them, especially if obedience is accompanied by prudence, yet the authority of its precepts as commands does not rest on this . Reason uses such facts only (by way of counsel) as a counterpoise to the temptations to the opposite, in order, first of all, to compensate the error of an unfair balance, so that it may then assure a due preponderance to the à priori grounds of a pure practical reason.

If, therefore, we give the name Metaphysic to a system of à priori knowledge derived from mere concepts, then a practical philosophy, which has for its object not nature but freedom of choice, will presuppose and require a metaphysic of morals: that is, to have it is itself a duty, and, moreover, every man has it in himself, though commonly only in an obscure way; for without à priori principles how could he believe that he has in him a universal law-giving? Moreover, just as in the metaphysic of natural philosophy there must be principles touching the application to objects of experience of those supreme universal laws of a physical system generally: so also a metaphysic of morals cannot dispense with similar principles; and we shall often have to take the special nature of man, which can only be known by experience, as our object, in order to exhibit in it the consequences of the universal moral principles; but this will not detract from the purity of the latter nor cast any doubt on their à priori origin—that is to say, a Metaphysic of Morals cannot be founded on anthropology, but may be applied to it.

The counterpart of a metaphysic of morals, namely, the second subdivision of practical philosophy generally, would be moral anthropology, which would contain the subjective conditions favourable and unfavourable to carrying out the laws of the power in human nature. It would treat of the production, the propagation, and strengthening of moral principles (in education, school and popular instruction) , and other like doctrines and precepts based on experience, which cannot be dispensed with, but which must not come before the metaphysic, nor be mixed with it. For to do so would be to run the risk of eliciting false, or at least indulgent moral laws, which would represent that as unattainable which has only not been attained because the law has not been discerned and proclaimed in its purity (the very thing in which its strength consists); or else because men make use of spurious or mixed motives to what is itself good and dutiful, and these allow no certain moral principles to remain; but this anthropology is not to be used as a standard of judgment, nor as a discipline of the mind in its obedience to duty; for the precept of duty must be given solely by pure reason à priori.

Now with respect to the division to which that just mentioned is subordinate, namely, the division of philosophy into theoretical and practical, I have explained myself sufficiently elsewhere (in the Critical Examination of the Faculty of Judgment), and have shown that the latter branch can be nothing else than moral philosophy. Everything practical which concerns what is possible according to physical laws (the proper business of Art) depends for its precept on the theory of physical nature; that only which is practical in accordance with laws of freedom can have principles that do not depend on any theory; for there can be no theory of that which transcends the properties of physical nature. Hence by the practical part of philosophy (co-ordinate with its theoretical part) we are to understand not any technical doctrine, but a morally practical doctrine; and if the habit of choice, according to laws of freedom, in contrast to physical laws, is here also to be called art, we must understand thereby such an art as would make a system of freedom like a system of nature possible; truly a divine art, were we in a condition to fulfil by means of reason the precepts of reason, and to carry its Ideal into actuality.